11 papers
An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models
Dengyang Jiang, Ruoyi Du, Zhennan Chen +10
This paper investigates an increasingly important topic in generative modeling: pixel-space diffusion models. Although numerous studies have explored this topic, most focus on smal…
From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models
Zanyi Wang, Xin Lin, Haodong Li +2
Large-scale text-to-image models are attractive backbones for dense prediction because RGB generation pretraining learns rich semantic, structural, and geometric priors. Existing g…
D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
Dengyang Jiang, Xin Jin, Dongyang Liu +9
The landscape of high-performance image generation models is currently shifting from the inefficient multi-step ones to the efficient few-step counterparts (e.g, Z-Image-Turbo and…
Exploring Time Conditioning in Diffusion Generative Models from Disjoint Noisy Data Manifolds
Liuzhuozheng Li, Zhiyuan Zhan, Shuhong Liu +5
Practically, training diffusion models typically requires explicit time conditioning to guide the network through the denoising sampling process. Especially in deterministic method…
Unlocking the Potential of Grounding DINO in Videos: Parameter-Efficient Adaptation for Limited-Data Spatial-Temporal Localization
Zanyi Wang, Fan Li, Dengyang Jiang +4
Spatio-temporal video grounding (STVG) aims to localize queried objects within dynamic video segments. Prevailing fully-trained approaches are notoriously data-hungry. However, gat…
Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning
Chengzu Li, Zanyi Wang, Jiaang Li +9
Vision-Language Models have excelled at textual reasoning, but they often struggle with fine-grained spatial understanding and continuous action planning, failing to simulate the d…
Distribution Matching Distillation Meets Reinforcement Learning
Dengyang Jiang, Dongyang Liu, Zanyi Wang +12
Distribution Matching Distillation (DMD) facilitates efficient inference by distilling multi-step diffusion models into few-step variants. Concurrently, Reinforcement Learning (RL)…
RefTon: Reference person shot assist virtual Try-on
Liuzhuozheng Li, Yue Gong, Shanyuan Liu +7
We introduce RefTon, a flux-based person-to-person virtual try-on framework that enhances garment realism through unpaired visual references. Unlike conventional approaches that re…
Deforming Videos to Masks: Flow Matching for Referring Video Segmentation
Zanyi Wang, Dengyang Jiang, Liuzhuozheng Li +6
Referring Video Object Segmentation (RVOS) requires segmenting specific objects in a video guided by a natural language description. The core challenge of RVOS is to anchor abstrac…
TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIP
Fan Li, Zanyi Wang, Zeyi Huang +3
3D visual grounding allows an embodied agent to understand visual information in real-world 3D environments based on human instructions, which is crucial for embodied intelligence.…
AffordanceSAM: Segment Anything Once More in Affordance Grounding
Dengyang Jiang, Zanyi Wang, Hengzhuang Li +7
Building a generalized affordance grounding model to identify actionable regions on objects is vital for real-world applications. Existing methods to train the model can be divided…